WO2023163248A1 - Server, method, and program that matches food ingredients on basis of artificial intelligence - Google Patents

Abstract

The present disclosure relates to a method that matches food ingredients on the basis of artificial intelligence, wherein the method may calculate component power for each test item for each component, calculate a component index that is the sum of the component powers for each test item for each component, and then derive optimized food ingredients for an organism on the basis of the calculated component index.

Description

AI-based ingredient matching server, method and program

The present disclosure relates to a food ingredient matching method.

Most of the drugs prescribed to a subject with a specific disease or a subject in need of improvement in health condition are often sourced from herbal medicines or components derived from living organisms.

In the case of providing a diet by matching ingredients optimized for the above-mentioned target using this point, it is expected that diseases can be rapidly improved or health conditions can be rapidly improved.

However, due to the nature of modern medicine, there is a problem that even a doctor may lack medical knowledge in other fields than his or her specialty, and it is too difficult to know all of the vast knowledge about food ingredients.

Conversely, in the case of a person who has extensive knowledge about food ingredients and nutrients, there is a problem in that it is too difficult to have the same medical knowledge as a doctor.

If the health condition of an organism is accurately analyzed and suitable ingredients are provided, it is expected to improve the health condition of the organism or improve the disease, but due to the above problems, such technology does not currently exist. .

The present disclosure for solving the problems described above is to provide an artificial intelligence-based ingredient matching method.

In addition, according to the present disclosure, component power for each test item is calculated for each component, component index, which is the sum of component powers for each test item, is calculated, and then food ingredients for organisms are derived based on the calculated component index. want to do

The problems to be solved by the present disclosure are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

An artificial intelligence-based food ingredient matching method according to an embodiment of the present disclosure for solving the above problems is a method performed by a service server, comprising: calculating component power for each test item with respect to a first component; calculating a first component index that is a sum of component powers for each test item for the first component; and deriving at least one food ingredient based on the first component index, wherein the component power calculation step includes: calculating a first weight corresponding to the importance of each item of the inspection item; calculating a second weight corresponding to the resultant value of the test item;

calculating a third weight based on the second weight; and calculating the component power by multiplying the first weight, the second weight, and the third weight.

In addition, the calculating of the second weight may include calculating the second weight according to the severity of the test result for each test item based on the result value.

In addition, the third weight is used to increase, decrease, or nullify the influence of the second weight by a specific coefficient, and in the calculating of the third weight, the type of the test item and the second weight calculated for the corresponding test item It is characterized in that the third weight is calculated based on.

In addition, the third weight is a weight for the efficacy of the first component on the test subject according to the result value of the test item, and the calculating of the third weight includes the type of the first component and the type of the corresponding test item. And the third weight may be calculated based on the result value of the corresponding inspection item.

In addition, the examination item includes at least one higher classification item, and the service server may calculate the first weight by multiplying a classification weight set for at least one higher classification item to which the examination item belongs.

In addition, the step of deriving the food material may include correcting the calculated first component index; Setting the number of extracted food ingredient data based on the corrected first component index; Extracting ingredients as many as the set ingredient data extraction number; Calculating a score for each ingredient by summing at least one first component index corresponding to each of the extracted ingredients; and providing a required intake amount of each food ingredient based on the calculated score for each food ingredient.

In addition, after the ingredient derivation step, a material post-processing step of deriving final ingredient information by performing at least one of filtering and alternative ingredient derivation for the derived ingredient may be further included.

In addition, after the food ingredient post-processing step, matching the food corresponding to the derived final ingredient information; and providing a diet based on the matched food.

In addition, the artificial intelligence-based food ingredient matching server according to an embodiment of the present disclosure for solving the above problems includes a memory in which a weight calculation algorithm is stored; and calculating component power for each test item for the first component, calculating a first component index that is the sum of component powers for each test item for the first component, and deriving at least one food ingredient based on the first component index. and a processor that calculates a first weight corresponding to the importance of each item of the inspection item, calculates a second weight corresponding to a result value of the inspection item, and calculates a second weight based on the second weight. A third weight may be calculated, and the component power may be calculated by multiplying the first weight, the second weight, and the third weight.

In addition to this, another method for implementing the present disclosure, another system, and a computer readable recording medium recording a computer program for executing the method may be further provided.

According to the present disclosure as described above, it is possible to provide an artificial intelligence-based ingredient matching method.

In addition, according to the present disclosure, after calculating the component power for each test item for each component and calculating the component index, which is the sum of the component powers for each test item for each component, an optimized food material for living organisms based on the calculated component index can be derived.

The effects of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a block diagram of an artificial intelligence-based ingredient matching system according to an embodiment of the present disclosure.

2 is a flowchart of an artificial intelligence-based ingredient matching method according to an embodiment of the present disclosure.

3 and 4 are diagrams illustrating types of inspection items.

5 is a detailed flowchart of the component power calculation step.

6 is a diagram illustrating section matching of a second weight and a third weight.

7 is a detailed flow chart for the step of deriving ingredients.

8 is a diagram illustrating extracting high-ranking ingredients by the number of ingredients or ingredient ratios from a food ingredient DB sorted based on ingredient content.

9 is a diagram illustrating a process following the ingredient derivation step.

10 is a diagram illustrating output data of an artificial intelligence-based ingredient matching method according to an embodiment of the present disclosure.

Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of the following embodiments taken in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments described below and can be implemented in various different forms, only these embodiments make the invention of the present invention complete, and the common practice in the art to which the present invention belongs. It is provided to fully inform the person skilled in the art of the scope of the invention, and the invention is only defined by the scope of the claims.

Terminology used herein is for describing the embodiments and is not intended to limit the present invention. In this specification, singular forms also include plural forms unless specifically stated otherwise in a phrase. As used herein, "comprises" and/or "comprising" does not exclude the presence or addition of one or more other elements other than the recited elements. Like reference numerals throughout the specification refer to like elements, and “and/or” includes each and every combination of one or more of the recited elements. Although "first", "second", etc. are used to describe various components, these components are not limited by these terms, of course. These terms are only used to distinguish one component from another. Accordingly, it goes without saying that the first element mentioned below may also be the second element within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used with meanings commonly understood by those skilled in the art to which the present invention belongs. In addition, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly specifically defined.

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

1 is a block diagram of an artificial intelligence-based ingredient matching system 10 according to an embodiment of the present disclosure.

The artificial intelligence-based ingredient matching system 10 according to an embodiment of the present disclosure includes a service server 100, a client terminal 50, and a medical staff terminal 70.

The service server 100 includes a processor 110, a communication unit 130, a memory 150, and an input/output unit 170.

However, in some embodiments, the ingredient matching system 10 or the service server 100 may include fewer or more components than those shown in FIG. 1 .

The AI-based ingredient matching device according to an embodiment of the present disclosure may be configured as a service server 100 of an AI-based ingredient matching server including a server device as shown in FIG. 1 .

The artificial intelligence-based food ingredient matching server according to an embodiment of the present disclosure generates and outputs output data according to an internal algorithm and process when input data for an analysis target is input.

In this case, any object of analysis can be applied if it is an organism.

As a representative example, it can be applied to a patient with a specific disease or a person in need of improvement in health condition, but is not limited thereto and can be applied to companion animals, animals in zoos, as well as plants.

Therefore, in the embodiments of the present disclosure, it is referred to as an analysis target.`

The communication unit 130 communicates with at least one of the client terminal 50 and the medical staff terminal 70, receives input data for an artificial intelligence-based ingredient matching service, and provides generated result data to the client terminal 50. can do.

The client terminal 50 is a device that requests analysis of the analysis target from the service server 100, and any information processing means such as a computer, smart phone, or tablet PC can be applied.

In addition, as described above, the analysis target for which analysis is requested from the client terminal 50 may be the client himself, but the client's family and acquaintances as well as animals and plants may be applied.

The medical staff terminal 70 is a device that provides medical data for an analysis target to the service server 100, and various examples such as a doctor terminal, a medical staff terminal, and a hospital server can be applied.

In some embodiments, when the service server 100 receives a request for a food ingredient matching service for an analysis target from the client terminal 50, it connects to a server of an affiliated medical institution to receive medical data for the analysis target, and automatically You can also input the algorithm and output the result.

The service server 100 may provide an AI-based ingredient matching service through a service application or web.

The memory 150 may store weight calculation algorithms, artificial intelligence models, details of food ingredient matching of analysis targets, and the like.

In addition, the memory 150 stores various commands, algorithms, etc. for providing an artificial intelligence-based ingredient matching service according to an embodiment of the present disclosure.

The input/output unit 170 may directly receive a control signal from the service server 100 or output information such as generated results and matching results.

In some embodiments, the input/output unit 170 can be used by the manager of the service server 100, and can also be used for algorithms, firmware updates, output of inspection results to check for errors, and the like.

The processor 110 controls the operation of components in the service server 100, and the artificial intelligence-based ingredient matching method according to an embodiment of the present disclosure using commands, algorithms, and artificial intelligence models stored in the memory 150 can provide.

Below, with reference to the flow chart of FIG. 2 and other exemplary drawings, the artificial intelligence-based ingredient matching server and method according to an embodiment of the present disclosure will be described in more detail.

2 is a flowchart of an artificial intelligence-based ingredient matching method according to an embodiment of the present disclosure.

3 and 4 are diagrams illustrating types of inspection items.

The processor 110 calculates component power for each test item with respect to the first component. (S100)

The processor 110 calculates a first component index that is the sum of component powers for each test item for the first component. (S200)

The processor 110 derives at least one food ingredient based on the first ingredient index. (S300)

The artificial intelligence-based ingredient matching server according to an embodiment of the present disclosure calculates ingredient power for each inspection item for various ingredients.

Therefore, the first component means one of various components such as calcium, magnesium, copper, vitamins, omega 3 fatty acids, methionine, cryptophane, selenium, zinc, gluten, etc. Finally, the service server 100 analyzes the analysis target In order to provide optimized ingredients, we analyze all ingredients.

The processor 110 calculates component power for each test item for a specific component through S100 and calculates the component index by summing the calculated component powers for all test items of the corresponding component.

This calculates the component power of which component is important and necessary to the analysis target according to the resultant value according to the input value of the analysis target for each test item. , because it may correspond to a plurality of inspection items.

For example, if the component power of vitamin C is calculated as 10 in the first test item and the component power of 20 is calculated in the second test item for the test target, this means that vitamin C is an important component for the component target. Therefore, it means to calculate 30 as the component index by adding 10 and 20.

When the calculation of the component index, which is the sum of component powers for each inspection item for all components, is completed through S100 and S200, the processor 110 proceeds to S300 to determine at least one component index for analysis target based on the component index calculated for all components. derive the ingredients.

Referring to FIG. 3 , the types of inspection items are exemplified, the inspection items include at least one level higher classification items, and the lowest classification item list in the classification items of FIG. 3 means each inspection item.

In an embodiment of the present disclosure, the highest classified inspection items include inspection data and non-inspection data items.

The test data item (second test item) includes a test item requiring a medical test, and includes a general test (clinical medical test) and a molecular biology test (basic medical precision test).

Referring to FIG. 3, general tests (clinical medical tests) include blood tests, feces, urine tests, otolaryngology tests, ophthalmology tests, pulmonary function tests, body composition analysis, CT scans, MRI scans, X-rays, bone density, ultrasound, and endoscopy. , electrocardiogram, and at least one of a biopsy, and in addition to those shown in FIG. 3, any test item corresponding to a general test (clinical medical test) may be included.

Referring to Figure 3, molecular biology test (basic medical precision test) is a cell metabolism test (urine), amino acid test (blood), NK cell activity test (blood), fatty acid test (blood), blood heavy metal, mineral test (blood ), active oxygen antioxidant test (blood), hair heavy metal, mineral test (hair), dementia test (blood), MAST allergy test (blood), lgG food reaction test (blood), lgG4 food reaction test (blood), cancer risk It includes at least one of a test (blood), a telomere genetic test (blood), and a genetic test (blood), and any test item corresponding to a molecular biology test (basic medical precision test) in addition to that shown in FIG. 3 may be included. there is.

Non-test data items include test items that do not require medical tests.

However, even if a medical examination is not required, the client or the analysis target cannot independently input the examination items included in the non-test data items.

Specifically, the non-test data item is a data item that the client can directly input without going through a doctor (third inspection item) and a data item that needs to be entered through a doctor (fourth inspection item). are classified

In this case, the data item (fourth examination item) that has undergone a doctor may mean data that requires a doctor's judgment/diagnosis.

The data item (fourth examination item) that has passed through the doctor can be classified into a medical history item and a diagnosis data item (physical examination).

Referring to FIG. 3, the medical history items include at least one of amenorrhea, alcoholism, tonsillitis, angina pectoris, venous insufficiency, glaucoma, influenza, measles, Parkinson's disease, sinusitis, tumor, and cancer, in addition to those shown in FIG. Any test item that corresponds to the medical history item may be included.

Referring to FIG. 3, the examination data items refer to examination items requiring a physical examination, and include biological symptoms, general appearance and behavior of the subject to be analyzed, body temperature, pulse rate, respiratory rate, blood pressure, overall inspection, overall palpation, complexion, rash, It includes at least one of intraocular pressure and strabismus, and any examination items corresponding to examination data items other than those shown in FIG. 3 may be included.

The data item (third inspection item) that has not passed through a doctor may include a simple personal information item of an analysis target and a query data item for the analysis target.

Referring to FIG. 3, the simple personal information item includes at least one of the name, contact information, gender, and date of birth of the subject to be analyzed, and any items corresponding to simple personal information other than those shown in FIG. 3 may be included.

The query data item may include a symptom item, a family history item, and a social history item.

The symptom items include at least one of fatigue, weight loss, fever, hair loss, pigmentation, headache, itching, cough, palpitations, constipation, dysmenorrhea, premenstrual syndrome, insomnia, and anxiety, symptoms in addition to those shown in FIG. Any inspection item corresponding to the item may be included.

The family history item includes at least one of hypertension, diabetes, hyperlipidemia, and cancer, and may include any test item corresponding to the family history item other than those shown in FIG. 3 .

The social history item includes at least one of pregnancy, smoking, constitution, vegan, religion, occupation, and nationality, and may include any test items corresponding to the social history item in addition to those shown in FIG. 3 .

The artificial intelligence-based food ingredient matching method according to an embodiment of the present disclosure derives ingredients based on input data for each inspection item for the analysis target, but does not use them as they are, but applies various weights to reflect and apply them, thereby optimizing the analysis target. results can be derived.

Hereinafter, an embodiment of weight calculation will be described in detail.

5 is a detailed flowchart of the component power calculation step (S100).

Referring to FIG. 5 , the component power calculation step ( S100 ) includes the following steps.

The processor 110 calculates a first weight corresponding to the importance of each item of the examination item. (S110)

Various inspection items have been described with reference to FIGS. 3 and 4 , and through this, it can be seen that the importance of each category to which the inspection item belongs is different.

The first weight is a category weight, and means a weight given by reflecting the diagnostic reliability of each examination method.

For example, the test data item may have a higher weight than the non-test data item and the doctored data (fourth test item), and the non-test data item and the doctored data (fourth test item) do not go through the doctor. It may have a higher weight than untested data (third inspection item).

However, this is only a simple example and is not necessarily limited in this way.

As described above with reference to FIGS. 3 and 4 , the inspection item includes at least one upper classification item.

And, each classification item has a classification weight set.

Referring to FIG. 4 , a first classification weight is set for a first inspection item (first classification weight) and a second inspection item (second classification weight).

In one embodiment, a third test item (2-1 classification weight), a 4th test item (2-2 classification weight), a general test item (2-3 classification weight), a molecular biology test item (2-2 classification weight) -4 classification weight) is set as the second classification weight.

As an embodiment, a simple personal information item (3-1 classification weight), a query data item (3-2 classification weight), a medical history item (3-3 classification weight), a diagnosis data item (3-4 classification weight) Classification weight) is a third classification weight set.

As an embodiment, a fourth classification weight is set for a symptom item (4-1 classification weight), a family history item (4-2 classification weight), and a social history item (4-3 classification weight).

In one embodiment, the inspection item may include at least one higher level classification item.

Further, the processor 110 may calculate a first weight by multiplying a classification weight set for at least one upper classification item to which the examination item belongs.

More specifically, the processor 110 may calculate the first weight by multiplying the classification weight set for all higher classification items to which each test item belongs.

For example, as shown in FIG. 4 , the first weight for social history is the result value of the analysis target for social history (the fourth classification weight for social history) X the third classification weight for the query data X the third test item. The second classification weight for X can be calculated based on the first classification weight for the first inspection item.

In this case, the fourth classification weight for social history means a result value calculated based on an input value of an analysis target for a corresponding test item.

In detail, the first weight for social history is based on the first classification weight X the 2-1 classification weight X the 3-2 classification weight X the result value (4-3 classification weight) of the analysis target for social history. can be calculated as

The processor 110 calculates a second weight corresponding to the result value according to the input value of the test item. (S130)

The second weight is a weight assigned according to the severity (severity) of the test result, and is assigned for each numerical range of data values, and the numerical range may be different depending on the type of data.

For example, test item A may include 4 numerical ranges, test item B may include 5 numerical ranges, and test item C may include 7 numerical ranges.

The second weight does not necessarily have to be weighted, and minus, 0, and plus can all be applied, and a decimal point value can also be applied.

As a first embodiment, the processor 110 may calculate the second weight so that the test value (result value output through the input value of the test target) is given according to the degree of deviation from a preset reference value.

As a second embodiment, the processor 110 may calculate the second weight so that the test value (result value output through the input value for the test subject) is assigned to each grade to which it belongs. (e.g., very severe/severe/mild or extremely insufficient/slightly insufficient, etc.)

The processor 110 calculates a third weight based on the second weight calculated in S130. (S150)

The third weight may have the following purpose.

① The third weight may increase, decrease, or nullify the influence of the second weight by a specific coefficient.

② The third weight may be for a weight for the efficacy of the first component on the test subject according to the output data of the test item.

The processor 110 may calculate a third weight based on the type of the corresponding examination item and the second weight calculated for the corresponding examination item.

More specifically, it has been mentioned that the second weight for the test item includes at least one numerical range.

6 is a diagram illustrating section matching of a second weight and a third weight.

When calculating the third weight, the processor 110 may calculate the second weight by section matching.

For the same test item, the third weight may include the same number of numerical ranges as the second weight.

This is to increase, decrease, or invalidate the influence of the second weight by a specific coefficient, like the ① feature.

For example, the second weight may be set so that the first inspection item has four numerical ranges of 0/0/1/5 according to the range of result values.

In this case, the third weight of the first inspection item may be set to have the same four numerical ranges as the second weight.

However, only the number of numerical ranges of the second weight and the third weight may be the same, and the weight may be different.

7 is a detailed flow chart for the step of deriving ingredients.

8 is a diagram illustrating extracting high-ranking ingredients by the number of ingredients or ingredient ratios from a food ingredient DB sorted based on ingredient content.

Referring to FIG. 7 , the ingredient derivation step includes the following steps.

The processor 110 corrects the first component index calculated in S200. (S310)

In detail, the processor 110 may correct the component indices by adjusting the component indices of all components calculated in S200 into percentiles.

The processor 110 sets the number of ingredients data extraction. (S330)

The processor 110 extracts as many ingredients as the number of ingredient data extractions set in S330. (S350)

The processor 110 sets the number of extracted ingredient data based on the ingredient index corrected in S310, and the following embodiments may be applied.

① The processor 110 may extract food ingredient data as many as the number corresponding to the percentile number of the corrected ingredient index.

Referring to FIG. 8 , in the case of proceeding as in ①, the processor 110 may extract ingredients corresponding to the 1st to 17th ranks among the ingredients containing calcium from the ingredient DB.

② The processor 110 may extract food ingredient data by a ratio corresponding to the percentile number of the corrected ingredient index.

Referring to FIG. 8 , when proceeding as in ②, the processor 110 may extract ingredients corresponding to the top 17% among ingredients containing calcium from the ingredient DB.

The processor 110 calculates a score for each food ingredient by summing the first component indices corresponding to each food ingredient extracted in S350. (S370)

For example, if the example shown in FIG. 8 is lettuce, and lettuce is extracted as a top food ingredient in each of calcium and magnesium components, the score of lettuce is obtained by summing the component index for calcium (400) + the component index for magnesium (380). It can be calculated as 780.

In one embodiment, the processor 110 may assign a content weight for each ingredient.

In detail, the processor 110 may calculate the score for each food ingredient by multiplying the content ratio to the food ingredient corresponding to the first rank of the ingredient index.

For example, if 5 ingredients containing calcium are extracted, and the 1st to 5th ranked ingredients contain 50mg, 40mg, 30mg, 20mg, and 10mg of calcium per 100g, respectively, the processor 110 scores the first ingredient = 400, 2nd ingredient score = 400x0.8 = 420, 3rd ingredient score = 400x0.6 = 240, 4th ingredient score = 400x0.4 = 160, 5th ingredient score = 400x0.2 = 80. can do.

The processor 110 calculates the required intake amount of each food ingredient based on the score for each food ingredient calculated in S370. (S390)

In detail, the processor 110 calculates the required intake amount of each food ingredient based on the score for each food ingredient calculated in S370, but may derive ingredient information for one serving of the analysis target.

Alternatively, the processor 110 may calculate the required intake amount of each food ingredient based on the score for each food ingredient calculated in S370, but calculate the required intake amount in a specific weight unit to provide ingredient information. (e.g. per 100g)

9 is a diagram illustrating a process following the ingredient derivation step.

Referring to FIG. 9 , the following steps may be further included after the ingredient derivation step (S300).

The processor 110 proceeds with a post-processing process for the ingredients derived in S300. (S400)

The processor 110 matches the food based on the food material post-processed in S400. (S500)

The processor 110 provides a menu based on the food matched in S500. (S600)

In detail, the processor 110 performs a food ingredient post-processing process of deriving final ingredient information by performing at least one of filtering and alternative ingredient derivation for the ingredient derived in S300.

The memory 150 stores at least one of a first exceptional condition set in advance, a second exceptional condition set for an analysis target, and a special weight set for symptoms or diseases.

The processor 110 may determine the type of food ingredient by applying at least one of the first exceptional condition, the second exceptional condition, and a special weight.

The first exception condition includes the acquisition difficulty of each ingredient.

In this case, the acquisition difficulty may include an acquisition difficulty in comparison to weight, a price per weight, and the like.

The second exceptional condition includes at least one of food allergy and food propensity of the analysis target.

The memory 150 stores alternative ingredient information for ingredients having an acquisition difficulty equal to or higher than a predetermined difficulty level.

For example, in the case of ingredients that are initially too expensive or ingredients that are difficult to obtain in the user's country or location, the processor 110 may filter and replace the ingredients with alternative ingredients.

The processor 110 may determine an alternative food material set for the food material as the food material to be analyzed when the acquisition difficulty of the food material determined for the analysis target exceeds a predetermined difficulty level.

10 is a diagram illustrating output data of an artificial intelligence-based ingredient matching method according to an embodiment of the present disclosure.

Referring to FIG. 10, the artificial intelligence-based food ingredient matching method and apparatus according to an embodiment of the present disclosure not only provides information on the types and importance of ingredients to be supplemented by the analysis target, but also the types and importance of ingredients to be avoided. can provide.

In addition, the artificial intelligence-based food ingredient matching method and apparatus according to an embodiment of the present disclosure have an effect of providing ingredients suitable for an analysis target and providing food as well as recipes for menu meals.

The method according to an embodiment of the present disclosure described above may be implemented as a program (or application) to be executed in combination with a server, which is hardware, and stored in a medium.

The aforementioned program is C, C++, JAVA, machine language, etc. It may include a code coded in a computer language of. These codes may include functional codes related to functions defining necessary functions for executing the methods, and include control codes related to execution procedures necessary for the processor of the computer to execute the functions according to a predetermined procedure. can do. In addition, these codes may further include memory reference related codes for which location (address address) of the computer's internal or external memory should be referenced for additional information or media required for the computer's processor to execute the functions. there is. In addition, when the processor of the computer needs to communicate with any other remote computer or server in order to execute the functions, the code uses the computer's communication module to determine how to communicate with any other remote computer or server. It may further include communication-related codes for whether to communicate, what kind of information or media to transmit/receive during communication, and the like.

The storage medium is not a medium that stores data for a short moment, such as a register, cache, or memory, but a medium that stores data semi-permanently and is readable by a device. Specifically, examples of the storage medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc., but are not limited thereto. That is, the program may be stored in various recording media on various servers accessible by the computer or various recording media on the user's computer. In addition, the medium may be distributed to computer systems connected by a network, and computer readable codes may be stored in a distributed manner.

Steps of a method or algorithm described in connection with an embodiment of the present disclosure may be implemented directly in hardware, implemented in a software module executed by hardware, or a combination thereof. A software module may include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any form of computer readable recording medium well known in the art to which this disclosure pertains.

Although the embodiments of the present disclosure have been described with reference to the accompanying drawings, those skilled in the art to which the present disclosure pertains can be implemented in other specific forms without changing the technical spirit or essential features of the present disclosure. you will be able to understand Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

[Description of code]

10: AI-based ingredient matching system

100: service server

110: processor

130: communication department

150: memory

170: input/output unit

Claims (10)

1. As a method performed by the service server,

   Calculating component power for each test item with respect to the first component;

   calculating a first component index that is a sum of component powers for each test item for the first component; and

   Deriving at least one food ingredient based on the first component index; includes,

   In the component power calculation step,

   Calculating a first weight corresponding to the importance of each item of the inspection item;

   calculating a second weight corresponding to the resultant value of the test item;

   calculating a third weight based on the second weight; and

   Calculating the component power by multiplying the first weight, the second weight, and the third weight,

   AI-based ingredient matching method.
2. According to claim 1,

   In the second weight calculation step,

   Characterized in that the second weight is calculated according to the severity of the test result for each test item based on the result value,

   AI-based ingredient matching method.
3. According to claim 2,

   The third weight is for increasing, decreasing, or negating the influence of the second weight by a specific coefficient,

   In the third weight calculation step,

   Characterized in that the third weight is calculated based on the type of the corresponding test item and the second weight calculated for the corresponding test item,

   AI-based ingredient matching method.
4. According to claim 2,

   The third weight is a weight for the efficacy of the first component on the test subject according to the result value of the test item,

   In the third weight calculation step,

   Calculating the third weight based on the type of the first component, the type of the corresponding test item, and the result value of the corresponding test item,

   AI-based ingredient matching method.
5. According to claim 1,

   The inspection item includes at least one higher level classification item,

   The service server,

   Characterized in that the first weight is calculated by multiplying a classification weight set for at least one higher classification item to which the inspection item belongs,

   AI-based ingredient matching method.
6. According to claim 1,

   In the step of deriving the ingredients,

   correcting the calculated first component index;

   Setting the number of extracted food ingredient data based on the corrected first component index;

   Extracting ingredients as many as the set ingredient data extraction number;

   Calculating a score for each ingredient by summing at least one first component index corresponding to each of the extracted ingredients; and

   Based on the calculated score for each food ingredient, providing the required intake amount of each food ingredient,

   AI-based ingredient matching method.
7. According to claim 1,

   Following the step of deriving the ingredients,

   Further comprising a food ingredient post-processing step of deriving final ingredient information by performing at least one of filtering and deriving alternative ingredients for the derived ingredient,

   AI-based ingredient matching method.
8. According to claim 7,

   After the food material post-processing step,

   Matching the food corresponding to the derived final ingredient information; and

   Further comprising providing a diet based on the matched food,

   AI-based ingredient matching method.
9. a memory in which a weight calculation algorithm is stored; and

   Calculate component power for each test item for the first component;

   Calculating a first component index, which is the sum of component powers for each test item for the first component;

   A processor for deriving at least one food ingredient based on the first component index,

   the processor,

   A first weight corresponding to the importance of each item of the inspection item is calculated, a second weight corresponding to a result value of the inspection item is calculated, a third weight is calculated based on the second weight, and the first weight is calculated. Characterized in that the component power is calculated by multiplying the weight, the second weight, and the third weight.

   AI-based ingredient matching server.
10. A computer-readable recording medium in which a program for executing the method of any one of claims 1 to 8 is stored in combination with a computer, which is hardware.

Images